Difference between ext2, ext3 and ext4

Leave a reply

List of differences between ext2, ext3, and ext 4 Linux file systems. One of the Linux interview questions answered in this article!

Comparison : ext2 vs ext3 vs ext4

This is another Linux interview question. What is the difference between ext2, ext3, and ext4 file systems? Or Explain Linux file system ext2 vs ext3 vs ext4? In this article we will walk through these differences and lastly I will present you all of them in tabular format so that they are easy to quickly read during your preparations.

Let’s see each file system’s features and lastly their comparison with other file systems.

EXT2 file system

  • It’s a second extended file system that was created to overcome limitations of the EXT file system.
  • Introduced in 1993 by Remy Card. It was the first commercial-grade filesystem for Linux
  • Does not supports Journaling
  • Fit for SD cards & USB drives since it has high performance and low writes (as journaling is not available). USB and SD storage are limited with write cycles hence its best fit for them.
  • Limits: Individual file size 16GB to 2TB. File system size 2TB to 32TB.

Limits are calculated based on block size used.  Block size varies from 1KB to 8KB. For example,  If 1KB block size is used max file size can go up to 16GB and for 8KB it’s 2TB. Middle range sizes being 2KB and 4KB which has file size limits of 256GB & 2TB (not mentioned in above limits) respectively. The same applies to the File system size limits defined above.

EXT3 file system

  • It’s third extended file system was created to overcome limitations of the EXT2 file system.
  • Introduced in 2001 by Stephen Tweedie. It was the most common filesystem in any Linux distro.
  • Supports Journaling
  • Journaling keeps track of file changes which helps in fast recovery and reduce chances of data loss in case of a system crash
  • Limits: Individual file size 16GB to 2TB. File system size 4TB to 32TB.
  • Upgrading FS from ext2 to ext3 is an online process without downtime.

EXT4 file system

  • It’s the fourth extended file system that was created to overcome limitations of the EXT3 file system.
  • Introduced in 2008 by a team of developers. Its most the latest filesystem in ext family.
  • Supports Journaling
  • Lots of new features introduced. Extents, Backward compatibility, Persistent pre-allocation, Delayed allocation, Unlimited number of subdirectories, Journal checksum, Faster FS check, Transparent encryption.
  • Limits: Individual file size 16GB to 16TB. File system size up to 1EB.
  • Upgrading FS not needed. Due to backward compatibility, ext2, ext3 can be directly mounted as ext4.

All above points can be formatted in tabular format as below :

ParameterEXT2EXT3EXT4
Introduced year

199320012008
Developed by

 Remy Card Stephen Tweedie Team of developers
Journaling Not available Available Available
Individual file size

 16GB to 2TB 16GB to 2TB 16GB to 16TB
File system size

 2TB to 32TB 4TB to 32TB up to 1EB
Upgrade Can be done online to EXT3. Can be mounted as EXT4. No upgrade needed Can be mounted as EXT4. No upgrade needed NA

Get your Linux ISO download from these URLs

2 Replies

List of all Linux ISO download links. The latest and old releases of various Linux distro can be downloaded using links listed on this page.

Linux download links

Many times I come across questions from people or readers that where I can download RHEL iso? Where to download Ubuntu Linux? etc. Actually these download links are very easy to get from Google but still I thought of publishing them all in one place.

In this post, I will list all possible Linux download websites. Please note that all Linux distro’s server copy is taken into consideration for choosing download links.

  1. Red Hat Enterprise Linux RHEL latest release download link.  Old releases download link. You have to create a developer account here to get your ISO copy.
  2. CentOS Linux latest release download link. Old releases download link. No signup needed.
  3. Fedora Linux download link. No signup needed.
  4. Debian Linux download link. There are country wise FTP mirrors listed. Choose your nearest one and download it.
  5. Ubuntu Linux Latest release download link. Old releases download link
  6. Arch Linux download link. Country-wise HTTP mirror list.
  7. Kali Linux latest release download link. Old releases download link
  8. Linux mint latest release download link. Old releases download link
  9. OpenSUSE Linux latest release download link
  10. SUSE Linux 60 days trial download link
  11. Open Mandriva Linux download link
  12. Kubuntu Linux download link
  13. Bodhi Linux download link
  14. Gentoo Linux download link

The list will go on since Linux is open source and there are many distros floating around the internet! Report broken links and any additions in the comments below.

7 new features in RHEL7!

Leave a reply

Listing new features in RHEL7. These 7 new features making RHEL7 stand out from its predecessor. 

New features in RHEL7

Its been a while RHEL7 is launched and nicely accommodated in Linux world by now. What’s new in RHEL7? What is the difference between RHEL7 and its precedence versions? these kinds of questions are flowing through interviews these days. So thought of jotting them down. In this post I will quickly walk through some new key features launched in RHEL7 by Red Hat.

What’s new in RHEL7?

RHEL7 officially released in June 2014 with codename Maipo. It took time for the market to absorb this new release since there are many new features launched in this release. Red Hat launched many new ways, commands to do traditional stuff. We are going to see them now –

  1. The default file system is XFS. RHEL6 was launched with EXT4 as the default file system. XFS is a highly scalable, high-performance file system. XFS supports metadata journaling, which helps in quicker crash recovery. This means file system checks will take very little time. The XFS file system can be defragmented and extended while mounted. This makes it more admin friendly in a production environment since it avoids downtime of file systems for activities. Supports only 64 bit systems.
  2. Introduction of systemctl. A new way to manage services on RHEL7 is systemctl. Older service and chkconfig command is being replaced with systemctl.
  3. Run levels being called as targets. In RHEL7 run levels are called targets. Default target (run-level) is defined in /etc/systemd/system/default.target. 
  4. Fast boot. RHEL7 boots faster than its predecessors. This is achieved by simultaneously starting services that are not dependent on each other. In older versions, services used to start one after another. So if one service stuck or delays to start it subsequently delays the following process and boot time. This hurdle is removed in RHEL7 allowing it to boot much faster.
  5. New system service manager: systemd. The former init process is no more PID 1 or first process. Systemd is introduced which controls standard base init scripts.
  6. Bigger filesystem limits. RHEL7 now supports filesystem size up to 500TB. This limit is also the same for an individual file. This is due to the XFS file system on a 64-bit machine. RHEL6 supports the 16TB filesystem.
  7. New firewall identity. Former firewall iptables now replaces by firewalld (Firewall Dynamic). iptables still exist in the system and you can disable firewalld and use iptables.

These are key differences in the new RHEL7 release. apart from this there are many new things added in RHEL7. The whole list is compiled by Red Hat here.

Let us know your suggestions/feedback in comments section below!

How to unmount NFS when the server is offline

Leave a reply

Learn how to unmount NFS when the server is gone. Dead NFS mounts can be un-mounted using forceful and lazy umount command.

Unmount NFS share when server is gone

This article will help you to un-mount NFS share from the client when the NFS server is gone or offline or un-available or decommissioned. We have seen how to configure the NFS server and how to handle NFS stale file error. But what if your NFS server is gone and its shares are still mounted on clients. Normally, before shutting down the NFS server, all clients should be notified and advised to unmount NFS shares they are using from this server.

But, in case if any of the clients have still NFS mounted when the server goes down, then the client should forcefully un-mount it. Normal mount operation won’t be effective in such cases. When the NFS server is down, you observe below things on the client who has NFS share still mounted.

  1. df command hangs since it tries to fetch NFS information but the NFS server is not responding.
  2. Tools, utilities who use/check mount point information like Ignite backup shows below error. 
    NFS server xyz not responding still trying
  3. fuser command hangs when running for the NFS mount point.
  4. umount (normal) command fails with the below error. 
    root@kerneltalks # umount /data nfs umount: nfs_unmount: /data: is busy umount: return error 1.

In this case, you need to use forceful (-f switch) and lazy umount (-l switch) to un-mount this dead NFS mount point. Lazy un-mount detach the said mount point from file system tree and cleans its all references once it’s not busy anymore.

Lazy un-mount is available in most Linux distributions. If not, you should be fine with only forceful un-mount too. In HPUX lazy un-mount is not available.

root@kerneltalks # umount -f -l /data

Conclusion

You can identify dead NFS share by df, fuser commands that appear to hang, failing to umount command. Such a dead NFS mount point can be un-mounted using forceful and lazy umount command.

How to replay Linux session recorded by the script command

1 Reply

Learn how to replay the Linux session recorded by script command. Visual Linux session recording along with timing information plays past session recording in real-time.

Replay Linux session like video!

In our last article we learned how to record Linux sessions using the ‘script’ command. In this article we will walk through steps to replay recorded sessions by script command. Normally, script command saves recording in the plain text log file which can be viewed using cat, more, less, or vi commands. That would be only plain text having commands and their outputs in the order you executed them while recording.

If you want to view your recorded output as it is being played on the terminal you can do it using scriptreplay command. It will play your output just as you are typing it on the terminal! scriptreplay needs time logs as well to play recorded sessions. This time logs can be generated using –timing switch with the script command. Let’s walk through these steps.

How to record Linux session with timing

We will use script command with --timing switch followed by filename in which all timing logs will be saved.

[root@kerneltalks ~]#  script --timing=time.log capture.txt
Script started, file is capture.txt
[root@kerneltalks ~]# date
Thu Jul 27 02:42:46 EDT 2017
[root@kerneltalks ~]# hostname
kerneltalks
[root@kerneltalks ~]# echo "I love kerneltalks"
I love kerneltalks
[root@kerneltalks ~]# exit
exit
Script done, file is capture.txt

here we are saving timing information in time.log file and session recording in capture.txt file. Both are plain text files and can be viewed. If you look at time.log file :

[root@kerneltalks ~]# cat time.log
0.001666 53
0.009220 1
3.980549 1
0.103633 1
0.191978 1
0.096629 2
0.127128 75
0.000920 1
1.671676 1
0.143421 1
0.080831 1
0.152510 1
----- output trimmed----

It has two columns of data in it. The first column denotes the number of seconds elapsed after the last display action. The second column is the number of characters printed on the screen.

That’s it. You have recorded your Linux session along with timing logs. This recording (capture.txt) can be replayed using scriptreplay command since its timing information is available.

How to replay recorded Linux session

Now both logs timing and recording need to feed scriptreplay command to let the show begin! The format and switch would be the same. Command used to replay session will be :

# scriptreplay --timing=time.log capture.txt

To see it in action, I captured it in the GIF file below. See how to actually replay as if the user is typing in the terminal as it was at the time of recording!

‘scriptreplay’ command in action

It replays exactly with the same time difference between two commands as you did at the time of recording! It’s like watching what the user has done in his session in real-time. It’s more of a visual record of Linux session while script was textual records.

Record Linux session using the script command

Leave a reply

Learn how to record Linux sessions using script command. This will help to keep a record of your commands and their outputs printed on the terminal at the time of execution.

Record your Linux session using ‘script’ command

In our last article we saw how to save PuTTY session output in a file on your desktop. In this article we will learn how to record session output in a file on the Linux server itself. And yes, it’s recording. It can be replayed later at any time and view commands being run and output being shown as it was done in real-time at the time of recording.

There are two utilities used for this task. script command used to record session and scriptreply is used to replay the recorded session from the file. In this article we will see Linux session recording using script command. scriptreply command is covered in this article.

script command takes a filename as an argument without any switch. It will write your commands along with their outputs in this file. Remember only commands ran after script command will be recorded and till you stop recording by hitting cntl+d key combination or typing exit. Make sure you have proper file permissions on the log file you are asking script to write into. Also, make sure you have enough space for recording since if your outputs are pretty lengthy then logfile gonna grow big.

Let’s start with an example. Here we are recording output in myoutputs.txt file. Command will be script myoutputs.txt:

[root@kerneltalks ~]# script myoutputs.txt
Script started, file is myoutputs.txt
[root@kerneltalks ~]# date
Thu Jul 27 01:31:39 EDT 2017
[root@kerneltalks ~]# hostname
kerneltalks
[root@kerneltalks ~]# w
 01:31:46 up 1 min,  2 users,  load average: 0.49, 0.21, 0.08
USER     TTY      FROM             LOGIN@   IDLE   JCPU   PCPU WHAT
ec2-user pts/0    59.184.148.123   01:31    2.00s  0.00s  0.19s sshd: ec2-user [priv]
root     pts/1                     01:31    2.00s  0.01s  0.01s w
[root@kerneltalks ~]# exit
Script done, file is myoutputs.txt

Now, observe above output.

  1. After script command execution, it notifies you (Script started, file is myoutputs.txt) that it has started recording your session and log file where it is recording.
  2. Then I punched in few commands for testing like date, hostname, w.
  3. I stopped recording by hitting the cntl+d key combination (exit command works too). script stopped and informed (Script is done, the file is myoutputs.txt) it has stopped recording.

Lets look at logfile myoutputs.txt

# cat myoutputs.txt
     Script started on Thu 27 Jul 2017 01:31:35 AM EDT
     [root@kerneltalks ~]# date
     Thu Jul 27 01:31:39 EDT 2017
     [root@kerneltalks ~]# hostname
     kerneltalks
     [root@kerneltalks ~]# w
      01:31:46 up 1 min,  2 users,  load average: 0.49, 0.21, 0.08
     USER     TTY      FROM             LOGIN@   IDLE   JCPU   PCPU WHAT
     ec2-user pts/0    59.184.148.123   01:31    2.00s  0.00s  0.19s sshd: ec2-user [priv]
     root     pts/1                     01:31    2.00s  0.01s  0.01s w
     [root@kerneltalks ~]# exit

     Script done on Thu 27 Jul 2017 01:31:51 AM EDT

Voila! All commands (including shell prompt PS), their outputs are there in the logfile. Notice that it also added timestamps at the top and bottom of the logfile indicating when the recording was started and stopped.

How to append script command recording in same logfile

Every time script command runs, it empties its logfile and adds content to it. If you want to append your recording to previously recorded file then you need to use -a (append) switch with it. This will keep data in a log file as it is and add new data to the bottom of the file.

[root@kerneltalks ~]# script -a myoutputs.txt
Script started, file is myoutputs.txt
[root@kerneltalks ~]# echo " I love kerneltalks.com"
 I love kerneltalks.com
[root@kerneltalks ~]# exit
exit
Script done, file is myoutputs.txt

I used -a switch in the above example to append new recording in the same file we created earlier. Tested one echo command. Let’s check the logfile.

[root@kerneltalks ~]# cat myoutputs.txt
     Script started on Thu 27 Jul 2017 01:31:35 AM EDT
     [root@kerneltalks ~]# date
     Thu Jul 27 01:31:39 EDT 2017
     [root@kerneltalks ~]# hostname
     kerneltalks
     [root@kerneltalks ~]# w
      01:31:46 up 1 min,  2 users,  load average: 0.49, 0.21, 0.08
     USER     TTY      FROM             LOGIN@   IDLE   JCPU   PCPU WHAT
     ec2-user pts/0    59.184.148.123   01:31    2.00s  0.00s  0.19s sshd: ec2-user [priv]
     root     pts/1                     01:31    2.00s  0.01s  0.01s w
     [root@kerneltalks ~]# exit

     Script done on Thu 27 Jul 2017 01:31:51 AM EDT
     Script started on Thu 27 Jul 2017 01:41:13 AM EDT
     [root@kerneltalks ~]# echo " I love kerneltalks.com"
      I love kerneltalks.com
     [root@kerneltalks ~]# exit
     exit

     Script done on Thu 27 Jul 2017 01:41:24 AM EDT

You can see there are two recordings in the same log file. You can distinguish them by their start and stop time (highlighted in output)

Conclusion :

script command can be used to log commands and their outputs on the Linux server itself. This command can be used in user profiles to monitor user activity on a server provided logfile mount point has huge free space to accommodate data generated by users.

Happy Sysadmin Day 2017!

Leave a reply

Happy Sysadmin Day 2017! System Administrator Appreciation Day is always last Friday of July month yearly. This year it’s 18th annual.

Happy Sysadmin Day 2017!!

First of all,

Happy Sysadmin Day to all our readers

Yes, its 28 July 2017! System Administrator Appreciation Day 2017!

Sysadmin day is normally last Friday of July month. This year its 18th annual Sysadmin day.

Sysadmin! He/she is the one who keeps your digital world running from behind the scenes. Anything digital you use right from your computer to your ATM card machines, sysadmin is behind them to keep them running healthy. Sysadmin never gets any appreciation he deserves. He works 365 days of year-round the clock even in graveyard shifts. His clock always ticks to make services available all the time. When systems are down everyone else is at ease, the only sysadmin is the one who gets all the heat and overworked to get systems back up. You can not imagine the digital world without sysadmin!

What is sysadmin day

It is a day dedicated to the appreciating system administrator. They are the real hero keeping your digital world running! So why not dedicate one day to appreciate the work they do for you! Appreciate their hard work, their holiday sacrifices, their tolerance when they take heat during downtime, their extra workloads keeping your digital infra up to date. Let’s appreciate at least one sysadmin today by greeting them, gifting them, exchanging little talk with them & expressing thankfulness!!

There are few ways you can appreciate your sysadmin listed here on sysadminday.com. Those are really worth considering.

Cheers!!

Thank you readers (mostly all of you are sysadmins!) for stopping by! Enjoy your day! And yes, for my Linux buddies – Have a happy shell!

Out tweet about this post was liked by the Founder of Sysadmin day! Ted Kekatos @tedkekatos

 

How to save PuTTY session output automatically

2 Replies

Learn how to save PuTTY output in logfile locally on the desktop. Important settings to be done on PuTTY client before you start your sysadmin work!

Save PuTTY session output in file on desktop

PuTTY is one of the widely used SSH, telnet client to connect Linux Unix servers. Its open-source software and can be downloaded here. This article is intended to beginners who just stepped into the command-line world of *nix. Experienced sysadmins already knew what this article is all about!

Most of the beginners are not familiar with PuTTY settings and they just use default ones. This leaves them into two troubles :

  1. If the command output is too long, it scrolls up the PuTTY screen and the output beginning is not visible.
  2. After closing the PuTTY session, the user won’t have a trace of what he did in his session.

To avoid them we need to tweak few settings in PuTTY and set them in default ones. So that next time you fire up PuTTY client, it starts with tweaked settings as default and makes your life easy!

How to increase PuTTY scroll length

Many times your output is long enough to scroll beyond PuTTY screen buffer and you won’t be able to get to the beginning of the output. To avoid this you need to increase the screen buffer size in settings. Let’s see how to do it. Open PuTTY and select ‘Window‘ on left side pane as below :

Putty scrollback size

You will be presented with settings page on the right which has a field where you can set ‘Lines of scrollback‘. Enter the number of your choice here (big enough). That’s it. Now PuTTY will be having a scroll buffer of the number of lines you specified. Never lost your output beyond scroll!

How to save PuTTY session output on local storage

The next challenge is users don’t get any logs saved by default in PuTTY. So in case you want to a checklist of commands you ran and their outputs, it’s not possible to check (once you close your session). Here session logging comes to rescue. Let’s see how to set it. Open PuTTY and select ‘Logging‘ under ‘Session‘ on the left pane.

Save putty session logs

You will present with logging settings on right. Session logging is off by default. You need to check bullet box next to ‘All session output‘ to log all session output in file. Provide log file name using browse button (specify where to save logs on your system). You can use parameters like &Y, &M (meanings listed below option), so that every time you start a new session, new logfile have unique name. This will avoid overwriting existing file with same name.

PuTTY also offers you to set options what to do when it finds file name already existing on specified path. You can choose to overwrite, append or ask user.

Now whatever commands you type and outputs you see in your session will be saved as it is in a file on your computer as well. These log files can be opened in notepad/Wordpad to view later. Be sure you don’t leave your session running with commands like a top (which continuously changes data on the screen for infinite time). Because this will fill up your logfiles fast and you may end up filling your local hard disk!

Save customized settings in PuTTY as default settings

Now you learned these two settings which are very much important to set but you don’t want to set it everything you fire up PuTTY client. You can save them as default settings!

Open PuTTY and do the above changes. Now, click and select ‘Session‘ on left pane which will bring you back to the client home screen.

PuTTY default setting

Now click and select ‘Default Settings‘ as shown above and click ‘Save‘. That’s it! Your customized settings are now saved as default settings. Now whenever you start PuTTY client it will start with your customized settings.

Have any tips regarding PuTTY client? Share with us in the comments section below!

Rsync to EC2 linux server on AWS

4 Replies

Learn how to rsync to EC2 with the help of SSH protocol authenticated using a private key file. The process can be used for Rsync from and between EC2.

Rsync to EC2 Linux instance

We learned about Rsync in our last post. We learned how Rsync helps in a data backup or mirroring by using less bandwidth, time on the second run. Since it syncs only changes in later executions after the first fresh copy operation. Now many traditional data centers are moving to cloud services like AWS. Rsync can be useful to sync data from your local server to AWS hosted EC2 instance (if the data size is not huge).

In this article we will learn about how to rsync to EC2 server in AWS. Since you know EC2 Linux instances don’t use a conventional used id-password combination for authentication, Key pairs need to be used in Rsync for authentication EC2. For the Rsync setup, your EC2 instance must be launched with public-private key pair and you should have a private key file with you.

Get started

  • To start with making sure your EC2 instance is launched with a key pair.
  • Upload private key file on the source server (from where you are going to Rsync to EC2)
  • Make sure key file set with 400 permission
  • Get public IP or public DNS name of EC2 server from AWS EC2 console web page
  • Confirm you are able to connect from source to EC2. (verify AWS security groups and firewall settings)

Execute Rsync to EC2

We have testfile.tar for testing copy and private key file (mykey.pem) ready on the source server.

[root@kerneltalks ~]# ll /root/mykey.pem
-r--------. 1 root root 1675 Jul 24 01:01 /root/mykey.pem
[root@kerneltalks ~]# ll testfile.tar
-rw-r--r--. 1 root root 39198720 Dec 19  2016 testfile.tar

Now, use below Rsync command :

[root@kerneltalks ~]# rsync -avz -e "ssh -i /root/mykey.pem" testfile.tar ec2-user@ec2-13-126-114-120.ap-south-1.compute.amazonaws.com:/tmp/
sending incremental file list
testfile.tar

sent 8520069 bytes  received 31 bytes  3408040.00 bytes/sec
total size is 39198720  speedup is 4.60

Where –

  1. -a: Archive mode preserves permission and ownership
  2. -v: verbose mode
  3. -z: compress
  4. -e: Choose remote shell of execution
  5. ssh -i keyfile: Use the private key for authentication on destination using ssh protocol
  6. source (testfile.tar)
  7. Destination: Public DNS name of EC2 instance

That’s it! Your file is copied over to EC2. This can be done vice versa as well. You can sync files from the EC2 server to the local server as well. Just switch source-destination paths and you are all set to go.

Rsync between two EC2 servers

Rsync can be executed between two EC2 servers i.e. from one EC2 server to another. The same above command can be used. If you are doing it for EC2 instances within the same region then the Internal DNS name can be used in a command.

Conclusion :

Rsync is possible from, to, and between EC2 servers. Key file authenticated SSH protocol should be used in the Rsync command to achieve this.

How to transfer data between two EC2 Linux instances

Leave a reply

Small tutorial to learn how to transfer data between two EC2 Linux instances on AWS. Explains the use of key authentication in the SCP command.

Copy data between two EC2 Linux instances

EC2 instance in AWS is a server instance that uses key-based authentication for login. Now, beginners or first time EC2 users wonder how to copy files from one EC2 server to another? or how to transfer data between two EC2 instances? You can achieve it using key files in scp command.

Few pre-requisites are :

  • Source and destination EC2 instances should be in the same region (can be in different availability zones)
  • Both instances should be able to communicate over port 22. Configure security groups accordingly.
  • EC2 instances built with specifying key pairs at the time of launch.

Lets get into actual stuff now. For example consider below setup :

  1. Two EC2 servers in the same region installed with Red Hat Linux
  2. Security groups allow port 22 both direction communication
  3. Key pairs used at the time of launch for both servers.
  4. The private key file is ready with me (.pem file). It’s part of the key pair used during launch.

In this tutorial, we will be copying file testfile.tar from server kerneltalks1 to kerneltalks2. Look at the file on kerneltalks1.

[root@kerneltalks1 ~]# ls testfile.tar
-rw-r--r--. 1 root root 39198720 Dec 19  2016 testfile.tar

Now, you need to use secure copy scp command with key file for authentication like below :

[root@kerneltalks1 ~]# scp -i /root/mykey.pem testfile.tar ec2-user@172.31.24.59:/tmp

Here,

  • -i switch used to specify the key file
  • Then source file path
  • followed by destination in userid@hostname/ip:/path/ format.

You have to use the Private IP of the destination EC2 instance. You can get this from your AWS EC2 console web page. Here kerneltalks2 has private IP 172.31.24.59. You can use hostname if you make an entry in /etc/hosts file.  Remember different Linux distros on AWS have different login id to be used.

# scp -i /root/mykey.pem testfile.tar ec2-user@172.31.29.79:/tmp
The authenticity of host '172.31.29.79 (172.31.29.79)' can't be established.
RSA key fingerprint is 66:c4:ce:37:c6:e6:a1:6c:2f:f9:9b:f2:f5:05:e3:38.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added '172.31.29.79' (RSA) to the list of known hosts.
testfile.tar                                                                                                               100%   37MB  37.4MB/s   00:00

In the above output, you can see kerneltalks2 authenticated using key file and file transfer was completed.

[root@kerneltalks2 ~]# ls /tmp/testfile.tar
-rw-r--r--. 1 root root 39198720 Dec 19  2016 testfile.tar

Rsync is another good way to copy data between two EC2 instances. You can learn about it in our other article: Rsync for EC2 on AWS.

Conclusion :

Files and directories can be transferred between two ec2 instances using the same Linux scp command. Only the authentication part is to be done using a key file with -i switch.